The Impact of Game Customization and Control Mechanisms on Recall of Integral and Peripheral Brand Placements in Videogames
Frank E. Dardis and Mike Schmierbach
Pennsylvania State University
Anthony M. Limperos
University of Kentucky
As marketers invest more and more money into in-game brand placements, little research has tested the effects of videogame customization and controller type in relation to advertising effects, even though these factors have demonstrated importance in other areas of gaming research. Results from an experiment show that game customization significantly increases recall of an integral brand placement-one that is central to actual game play-but not of peripheral brands, which simply appear within the game. Regardless of brand type, players using a traditional controller exhibit significantly greater recall than those who use a newer, more naturally mapping controller. An interaction effect indicates that the influence of controller type disappears when customization is allowed; this effect is not specific to either type of brand. These results are interpreted through models of processing fluency and the limited capacity model of motivated mediated message processing. The article concludes with marketing implications regarding technological videogame advances.
Keywords: brand placement, videogames, in-game advertising, recall, customization
Video and computer game sales in the United States nearly tripled from 1996 to 2006, reaching $7.4 billion, and in 2004, sales of videogame items surpassed cinema box office receipts for the first time (Walsh, Kim, and Ross 2008). An increasing number of companies thus are investing in in-game ads, which let brands interact with potential consumers in a more refreshing and engaging manner than other forms of advertising (Clifford-Marsh 2009). Analysts predict that in-game advertising spending will increase faster than any other form, up to more than $800 million in 2012 compared with just $55 million in 2006 (Guadiosi 2008). The current gaming audience also appears to be increasingly made up of players who are physically active, early and obsessive product or service adopters, leaders in their social groups, and focused on the world beyond gaming (“Major League” 2009). Thus, the old stereotype of gaming as a “nerdy” outlet for techies has vanished, and advertisers and marketers increasingly are turning to this medium to reach extremely valuable consumers with innovative communication methods.
Although companies are investing larger amounts of money in their in-game brand communications, relatively little is known about the effectiveness of various types of advertising options available to marketers. Consider, for example, a typical videogame that provides multiple channels for a brand message. The most common include “traditional” display ads of various sizes that appear in different locations throughout the game. But games also allow for much more sophisticated brand communications in which the branded products actually become part of the game itself (e.g., the car a player uses during the driving game). Furthermore, many games offer players an opportunity to customize or personalize their gaming experience by determining in-game properties, which can enhance marketing outcomes (Kwak et al. 2010). Some research on game customization has focused on usability (Pinelle, Wong, and Stach 2008), learning (Gee 2005), motivations (Yee 2006), and physiological arousal (Bailey, Wise, and Bolls 2009), yet few studies have considered the impact of customization on in-game advertising. Therefore, we incorporate literature on product placement, product integration, and processing fluency (Hang and Auty 2011) to investigate the impact of game customization on recall of in-game brand placements.
Brands also might attempt to get their names involved with technological advancements as they arise in the videogame industry (e.g., the newest games, control systems), lest they be left out of a dynamic new trend. For example, since its initial release, Nintendo Wii has dominated the console game market, outpacing both Microsoft’s Xbox 360 and Sony’s PlayStation 3 in terms of total global sales and market share (Parker 2010). The Wii’s success has been attributed largely to its nontraditional, motion-sensing controller (Hartley 2007). To keep pace with the Wii, both Microsoft and Sony have developed their own motion-sensing game controllers, in an attempt to entice and attract new gamers. This move toward motion controls in the console game market is likely to add another layer of complexity for advertisers who seek to place their brands within games. Diverse control schemes might be novel and attract more gamers to a particular game console, but this shift also might influence the effectiveness of in-game brand placements. For advertisers, simply throwing more marketing money at the latest gaming efforts and technologies cannot sufficiently address how well brand messages are working, if at all.
Although many types of game mechanisms appear in prior videogame research, we are not aware of any previous advertising study that has compared two control systems within the same experiment. Studies in areas other than advertising have compared different platforms or systems, with mixed results regarding the effects of changing technology (e.g., Ivory and Kalyanaraman 2007; Persky and Blaskovich 2007; Skalski et al. 2007). However, these studies focus on responses to the game, not the advertising contained therein. To rectify that omission, we apply the limited capacity model of motivated mediated message processing (LC4MP) (Lang 2006), which stipulates that the human mind has a finite amount of cognitive resources available to attend to message-processing tasks (e.g., remembering an in-game brand message). Accordingly, we investigate whether new videogame controllers differ from traditional ones in terms of their effects on ad memory.
In turn, this study aims to provide insight into several questions related to the effectiveness of in-game brand communications, along two important-and currently underinvestigated-dimensions: game customization and controller type. Both factors possess theoretical and practical merit. Any knowledge gleaned from this study therefore should help in-game marketers select with greater precision the gaming systems and/or genres and titles in which they should advertise. Furthermore, because in-game advertising, in both practice and research, is a relatively nascent field, these results should serve to assimilate extant videogame and advertising knowledge further, which could help shape future academic inquiries in this realm.
Literature Review
Integral and Peripheral Brand Placement in Videogames
Research on videogame advertising usually proposes that brand placements in games are more akin to product placement efforts than to traditional advertising messages. Both in-game placements and traditional product placements (e.g., in television shows or movies) are planned and unobtrusive messages (Balasubramanian 1994) that operate within the entertainment content (Karrh 1998) and include brand identifiers, in return for monetary or commercial considerations (Lee and Faber 2007; Schneider and Cornwell 2005). However, placements within videogames offer benefits beyond those in traditional media: They provide more specific audience targeting, a longer shelf life, and greater user involvement, because of the active nature of game playing versus the passive nature of absorbing other media content (Schneider and Cornwell 2005). This latter aspect of user involvement, or immersion (Nicovich 2005), best informs the conceptualization of brand placements (integral versus peripheral) that we adopt for this study.
By incorporating product placement research (e.g., Karrh 1998; Karrh, McKee, and Pardun 2003), Nelson (2002) and her colleagues (Nelson, Yaros, and Keum 2006) essentially initiated experimental research into the effectiveness of brand placements in auto-racing videogames. In both studies, the brand name placements names were “passive” billboard/sideboard ads that essentially provided a background to the broader racetrack context. Subsequent studies incorporated brand prominence as a measurable factor. Using the rationale established in research on product placement in movies (Gupta and Lord 1998), Schneider and Cornwell (2005) operationalized ad placements in an auto-racing game as low, medium, or high in prominence, depending on their size, color, position, and attractiveness. Their results demonstrated that prominent ads generated more recall and recognition than subtle ads. In another experiment to manipulate the location (focal versus peripheral) of ads in an auto-racing game, Lee and Faber (2007) found that focal product placement increased recall and recognition more so than peripheral placement, though the recognition superiority of focal placements disappeared when experienced players’ involvement in the game was high.
Perhaps the most interesting thing about these conceptualizations of brand placements is that they refer to the physical display components of the ads-mainly size and location within the game. Despite their revealing insights, these studies ignore another useful conceptualization of brands communication within videogames. In their study of movies, Gupta and Lord (1998) note that, in addition to larger physical qualities, prominent brands are those that are central to the narrative. In this sense, the term “product integration” means an advertising method that goes beyond simple product placement. For example, rather than simply attempting to work the product into an existing script at opportune times, product integration implies that brands prearrange multilevel deals with producers, negotiating production costs and advertising dollars, so the brand becomes an integral part of the show-that is, “becoming the script” (Wenner 2004, p. 113).
We contend that this distinction is analogous to a videogame context, in which branded products become part of the gaming activity through players’ actual use of and interaction with brands (e.g., driving a particular make and model of racecar), and in contrast with a brand identifier that simply appears within a game setting (e.g., a display billboard on the side of a racetrack). Not only does this conceptualization recognize that videogames are more immersive and engaging than other media types, because it allows for user interaction (Nicovich 2005), but it also incorporates brands’ continued realization that interactive marketing efforts through media such as videogames may be more effective than traditional advertising (Clifford-Marsh 2009).
Taken together, the notions of greater user immersion with the game and increased interaction with the brand itself lead us to define brand placements that become part of the game during player usage as “integral” to the gaming experience. This conceptualization is markedly different from the prior definitions of “prominent” or “focal” brand placements, which have relied more on physical attributes. However, existing conceptualizations of “peripheral” brand placements in games remain appropriate in two domains: when the ads are not physically part of the primary onscreen action, and when the brands do not constitute part of the in-game action, such that they are also conceptually peripheral to the game. With this study, we attempt to discover if any of the potential effects of customization and controller type on recall of in-game brand messages differ with the type of brand placement.
Customization
Several scholars suggest customization in video games is an important, theoretically interesting trait. For example, education scholars frequently cite an ability to customize as a critical feature of “good” games that can facilitate learning (e.g., Gee 2005), though little empirical work has tested this assumption. Therefore, customization might facilitate memory of advertising as well. Customization is also important for the general success of an interface, in that it creates usability (Pinelle, Wong, and Stach 2008) and the potential for flow (Chen 2007). For marketing, in-game personalization also increases players’ enjoyment, playing time, and game repurchase intentions (Kwak et al. 2010). Although these factors should enhance player immersion and enjoyment, they might undercut memory if players become so absorbed in the game that they do not notice the advertising (Chaney, Lin, and Chaney 2004). Yee (2006) finds that an ability to customize serves as a motivation to play massively multiplayer games, which offers direct empirical evidence of the importance of this factor. Bailey, Wise, and Bolls (2009) also provide experimental evidence that the ability to customize an avatar in a snack food “advergame” evokes greater presence and physiological arousal. Unfortunately, these authors do not report any measures of product-relevant attitudes or awareness, though prior research has connected presence to ad recall (Nelson, Yaros, and Keum 2006) and in-game ad evaluations (Nicovich 2005).
A recent study (Hang and Auty 2011) also has indicated that simply allowing a 10-second opportunity for players to select the integral brand used in a videogame (e.g., Nike shirts worn during a soccer match) increases recall of the in-game brand. In an important finding for our study, the group that both interacted with the game before playing it and played with Nike shirts demonstrated a higher recall than not only other groups that did not play with any branded shirts (which seems intuitive), but also another group that played the full game with Nike shirts but did not have an opportunity to customize or interact with the game prior to playing it.
Several studies (e.g., Lee and Labroo 2004; Shapiro 1999; Tulving and Schacter 1990; Whittlesea 1993) help explain this finding by noting two levels of processing fluency, which refers to the ease of stimulus processing and its effects on subsequent decisions. The perceptual level entails processing visual elements of stimuli, whereas the conceptual level relates to processing the language or other information communicated by stimuli. Hang and Auty (2011) contend that the differences in their study reflected synergistic effects of interacting with and playing the game with the Nike logo, which enhanced processing of both the physical stimuli (perceptual fluency) and conceptual-level features (brand). A final important, methodological note is that these authors also operationalize brand recall as the only measure of memory, because it avoids a cue-based stimulus test that is more appropriate for measuring perceptual fluency. Because we investigate the effects of both game customization and brand usage during the game on brand memory (i.e., enhanced conceptual fluency), we suggest that free recall is an appropriate memory measure.
Regarding brand placements, game customization seems most likely to influence the attributes related to integral, rather than peripheral, brands. Because customization increases general game immersion, and integral brand placement produces greater player-brand engagement, it seems likely that allowing players to customize the brand they use in a game leads to more interactivity with, and thus recall of, the brand than in the case when brands simply happen to appear in the backdrop or periphery of the game. Players who do not have an opportunity to customize the integral brand they use during the game do not experience as much interactivity or connectedness with the brand and accordingly should not remember the brand as well. Therefore, we propose:
H1: For an integral brand, ad recall is higher among players who customize their gaming experience than among players who do not customize.
Control Mechanism
Most videogame advertising research uses auto-racing games as their stimuli setting and relies on traditional control mechanisms (i.e., handheld or computer key/mouse), such that players’ fingers direct the flow of the game (e.g., Lee and Faber 2007; Nelson, Yaros, and Keum 2006; Schneider and Cornwell 2005). Nelson (2002) investigates both traditional, finger-based control mechanisms and a more realistic, naturally mapping (e.g., Skalski et al. 2007) steering wheel mechanism with accompanying foot pedals. However, that study did not compare the types of controllers in the same experiment; the two mechanisms simply added external support to the hypotheses results.
To the best of our knowledge, no studies have tested the effects of different control mechanisms on players’ ad recall within the same experiment, though several have investigated the impacts of differing systems on other player attributes. With a few exceptions (Limperos et al. 2011; Skalski et al. 2007; Tamborini and Skalski 2006), little empirical research has explicitly compared players’ enjoyment of popular games across control systems or technologies. Other research indicates that players prefer an immersive virtual environment system rather than a traditional game system (Persky and Blaskovich 2007) and that changing the technology of a driving simulator to increase the field of view (i.e., making it more “realistic”) enhances presence but does not increase enjoyment (Lin et al. 2002). In a study focused more specifically on actual control mechanisms, Skalski et al. (2007) find that controllers that afford more natural mapping during game play (e.g., steering wheels) enhance feelings of naturalness, presence, and enjoyment, compared with traditional controllers. In a similar study, Lindley, Le Couteur, and Bianchi-Berthouze (2008) compare players who used a bongo-drum controller or a more traditional controller to play a rhythm game. The drum players reported greater engagement, because the drum controller “afforded natural body movements” (Lindley, Le Couteur, and Bianchi-Berthouze 2008, p. 513), which allowed for more gestures that could facilitate social interactions.
Regarding the impact of controller type on ad memory, some direction comes from the LC4MP (Lang 2000, 2006), which is based on a limited capacity model of attention. Lee and Faber (2007) use the latter to investigate recall of in-game placements in general, without considering controller type. They refer to cognitive psychology (e.g., Kahneman 1973; Lynch and Srull 1982) to explain why cognitive attention is always limited: It must be allocated across two dimensions at any given time, namely, a primary task at hand and any other secondary tasks.
Building on this notion, the LC4MP predicts that a finite amount of cognitive resources can be allocated to a particular task and is specifically appropriate to study the effects of in-game brand placements. The model further holds that people allocate cognitive resources through specific mechanisms initially to encode a certain communication task (Lang, Potter, and Bolls 1999); environmental elements also automatically draw resources (Lang 2006). If more cognitive resources are allocated to a communication task than are required, the remaining available resources can be channeled toward encoding the information and beginning a three-step memory process. However, if the cognitive resources required outweigh the resources allocated (e.g., simultaneously performing other distracting tasks), the distracting task automatically diverts resources away from the initial encoding process and ultimately diminishes memory capabilities.
The three steps in the memory process in this model depend differentially on cognitive resource allocations: encoding, storage, and retrieval (Lang 2006). The first two elements, which are functions of reaction and association, respectively, can be gauged by measures of recognition and cued recall, respectively. The final process, information retrieval, is purely a function of memory and can be measured as free recall (Lang 2006). Because we focus on final, memory-based outcomes of in-game brand placements, we employ free recall as an appropriate gauge of players’ sensitivity to in-game brand messages.
The logic of LC4MP also applies to controller types. Traditional, handheld game controllers with which players are familiar should not require too much primary cognitive resource allocation and therefore should leave more resources available to attend to and process in-game brand placements. In contrast, newer control mechanisms, such as a steering wheel and foot pedals-even if they map more naturally or are fun for players-could demand more resource allocation as players learn the “feel” of the game. As a result, these players have fewer resources available to attend to brand messages, and they should demonstrate lower recall. No indications in prior research suggest whether the type of brand (integral or peripheral) should be affected by controller type. Therefore, we propose:
H2: Ad recall is higher for players who use a traditional controller rather than a newer, more technologically advanced controller.
From this review, we determine that both the control mechanism and customization should have effects on players’ cognitive, affective, and physiological states during game play. Because this study is the first to test the two factors within the same experiment, we propose the following research questions as well:
RQ1: Do game customization and controller type interact in their effects on ad recall?
RQ2: Do any effects of game customization or controller type vary according to whether the brand is integral or peripheral?
Method
A total of 96 undergraduate students were recruited from several large university classes and received extra credit for their participation. In the sample, 74% were women (n = 71) and 26% were men (n = 25), with an average age of 20.14 years. As control variables, neither gender nor age affected any of the results described in the next section. When they arrived at the lab, participants first completed a pencil-and-paper questionnaire to assess their prior experience with specific videogame titles/genres and the gratifications they sought from playing videogames (as part of another study). After completing the questionnaire, they were led into another lab room and told they would play an auto-racing game: Need for Speed on the Xbox 360 game platform. The lab room contained a large television, in front of which each participant sat and on which they completed some prestimulus tasks, then executed the actual racing game. Depending on the assigned condition, the rest of the procedure differed slightly.
Procedure
We employed a 2 (controller type: wheel versus traditional controller) ´ 2 (car customization: yes versus no) between-subjects design. All participants were randomly assigned to one of the four possible treatment conditions. Regardless of condition, all participants used the same racecar: the Mitsubishi Lancer EVO. The consistent, central exposure to this car served as the operationalization of integral brand placement. We preset this car’s specific performance levels (e.g., engine size, tire size) to ensure that all participants engaged in the same level of play during the game.
Those participants assigned to a customization condition immediately received a brief tutorial on how to customize their car before the racing game began. It showed them how to select basic aesthetic features for the car (e.g., paint color, numbers and decals, window tint color), without allowing them to access the menu to change the performance-level specifications. They were then left alone in the room for approximately five minutes to customize the car as they wished. Participants assigned to a noncustomization condition began by viewing an approximately five-minute slide show of photos of 26 cars, most of which were from car shows and similar events that attract highly customized cars. The photos did not indicate any action. These participants received a pencil-and-paper form and provided their opinion (1 = “don’t like”; 7 = “really like”) of each car as it appeared on the screen. The responses were not important; it was simply a five-minute exercise to get the noncustomization participants thinking about the aesthetics of cars, similarly conceptually to what participants in the customization condition were experiencing. In all the customization selections and the photos in the noncustomization condition, we confirmed that no advertisements appeared.
After completing these prestimulus tasks, participants were told they were about to play the racing game. Customization participants used the car they had customized; noncustomization participants were assigned to the default model of the car. Regarding controller type, half the participants used an actual plastic racing wheel made especially for Xbox 360, complete with gas and brake pedals on the floor, to control the car’s speed and steering direction during the race. The traditional controller condition instead employed the more common, handheld device, such that participants pushed various buttons and arrows with their fingers to control the car. Regardless of the controller, participants were instructed to complete two solo laps around a practice track, which would be the same track for the actual race. After the practice runs, a researcher selected from the onscreen menu the specific race track and skill level, both of which remained constant across all treatment conditions.
The researcher then explained that the playing session would consist of two complete races, each four laps long with seven other cars on the track. Participants were left alone in the room to complete the first race. The researcher then reentered the room, noted the participant’s race time, and restarted the game to run the second race on the same course. After the second race, we again noted the race time, and participants were led out of the room to complete a pencil-and-paper, post-stimulus questionnaire. The average time for the first race was 3 minutes, 39 seconds, and the average time for the second race was 3 minutes, 31 seconds. Race times did not vary across treatment conditions (p > .10).
Because it was the integral brand, players had to use the Mitsubishi car to participate in the game. Throughout the race, the screen presented a typical racing-game view of the back of the car, facing forward. The name Mitsubishi did not appear during game play, but the trunk of the car featured the brand’s three-diamond logo throughout the race. Participants were exposed to peripheral advertisements slightly prior to and during the actual race. The advertisements prior to the beginning of the race included display signage in a “pre-stage/show room” in which the racecar was waiting before the race (i.e., prior to the researcher selecting the proper race setting from the onscreen menu options, which typically took just a few seconds). Peripheral brand placements during the race included typical billboard/banner signage along the outside wall of the track and other realistic promotional materials (e.g., a large inflatable Castrol Syntec can on the infield). Advertisers from the showroom scene also appeared during the race. The in-race advertising did not change from lap to lap, so all participants were exposed to the same brands, number, and types of ads during both four-lap races.
Dependent Measure
To allow participants some “mind-clearing” time before asking them about brand recall, the relevant measure appeared in the post-stimulus questionnaire, which first gauged several other attitudinal variables related to participants’ gaming experience. The entire post-stimulus questionnaire took approximately 15 minutes to complete. The free recall measure instructed participants to “please list any brands or products that you remember seeing any ads for during your gaming experience.” For each of the four brands advertised during the game (Mitsubishi, Energizer, Progressive, Castrol), recall was entered as a dichotomous variable equal to 1 if participants listed the brand and 0 if they did not (see Nelson, Yaros, and Keum 2006). Therefore, the highest possible total recall score was 4. For the applicable analyses, we operationalized integral and peripheral recall separately, as we describe subsequently.
Additional Measures
In addition to gender, we assessed several other variables that could affect the results. However, the following preexisting variables did not differ across treatment conditions, nor did they have any effect on the results when introduced as covariates: age, gender, prior experience with the Need for Speed videogame on any system, and prior experience with the Xbox 360 system. Thus, we excluded these factors from the final analyses. Furthermore, the following measures of actual in-lab game play did not differ by condition, did not significantly predict ad recall, and did not alter the results of the analyses when introduced as covariates: enjoyment, perceived control, challenge-skill balance (not specific to controller type), concentration, transportation, and presence. In summary, we can safely rule out a variety of potential confounds and competing explanations for the results.
Results
We present the descriptive statistics in Table 1. The mean overall score for total recall was .39 (SD = .57), and approximately 28% (n = 27) of players recalled the integral brand (Mitsubishi), whereas only 8.3% (n = 8) remembered seeing any of the peripheral brands. In line with our prediction that ad recall for an integral brand is higher among players who customize their gaming experience, our 2 ´ 2 chi-square analysis revealed that participants who customized their car prior to game play were more likely to recall the Mitsubishi brand than were participants who did not customize (χ2(1) = 7.28, p < .01). Although the majority of players in each condition failed to recall seeing the brand, 40.0% (n = 20) of players in the customization conditions recalled seeing Mitsubishi, compared with only 15.2% (n = 7) of players in the noncustomization conditions. These finding confirm H1.
Table 1. Descriptive Statistics of Total Ad Recall by Treatment Condition
|
Condition |
Number |
Mean |
Standard Deviation |
|
Traditional Controller |
|||
|
Customization |
25 |
.44 |
.51 |
|
No customization |
23 |
.57 |
.79 |
|
Steering Wheel |
|||
|
Customization |
25 |
.44 |
.51 |
|
No customization |
23 |
.09 |
.29 |
We also predicted that total ad recall should be higher for players who use traditional rather than new controllers. A univariate analysis of variance (ANOVA) confirmed that players using the traditional, handheld controller recalled more ads (M = .50) than players who used the wheel (M = .27; F(1,94) = 4.02, p < .05, η2p = .04), in support of H2.
In RQ1 we considered the potential interaction effects of game customization and controller type on participants’ total recall. A univariate ANOVA revealed a significant interaction effect (F(1,91) = 3.97, p < .05, η2p = .04) that even overwhelmed the previously described effects. The mean total recall score (M = .44) was identical in the customization conditions across both controller types. However, when car customization was not allowed, participants who used the traditional controller (M = .56, n = 23) demonstrated significantly higher ad recall than did those who used the wheel (M = .09, n = 23) (see Figure 1). Their recall score also was higher than those of either customization group.
Figure 1. Interaction Effects of Game Customization and Controller Type on Total Ad Recall.
With RQ2, we inquired whether any effects of game customization and controller type might vary, depending on whether the brand was integral or peripheral during game play. To investigate this question, we ran two separate logistic regression models. Before starting the analyses, we recoded recall of peripheral brands into a dichotomous variable that measured whether participants remembered seeing any non-integral brands (i.e., recall of one or more peripheral brands was coded 1, and recall of no peripheral brands was coded 0). The first model used recall of Mitsubishi (0 = no, 1 = yes) as the dependent variable, and the second featured the dichotomized recall of peripheral brands as the dependent measure. In each model, the predictors were game customization, controller type, and their interaction effect.
In the first model, game customization emerged as the lone significant predictor of successful recall of the Mitsubishi brand from the game (b = 2.69, p < .05). This result lent further credence to the notion that game customization is an important determinant of whether players remember seeing integral brands during game play, as we predicted by H1. In the second model, no factors emerged as significant predictors of recall of peripheral ads. In this case, the interaction effect of game customization and controller type did not demonstrate significance for predicting recall of either the integral or the peripheral brand placements. Thus, though the ANOVA for RQ1 demonstrated a significant overall interaction, it did not appear specific to either type of brand placement, nor was it strong enough to emerge in either of the individual tests. Therefore, the interaction, though present, is not unique to either brand type.
Our inquiry into RQ2 contributes to the pattern of findings we revealed previously. First, allowing game customization of the integral brand increases recall of that brand, but it does not increase memory of peripheral brands. Second, a traditional controller elicits greater total recall of in-game placements, but this effect does not vary by brand type. Third, traditional controllers have the greatest impact when game customization is not allowed, which demonstrates not only that traditional controllers afford more opportunity to attend mentally to in-game ads but also that allowing customization can obviate or override any controller effects for integral brands.
Discussion
This study provides the first test of the simultaneous impact of game customization and controller type on recall of in-game brand placements. Game customization is a key element in eliciting recall of integral placements, but it does not affect the propensity to recall peripheral brands. We presume this finding is due to the increased immersion (along two dimensions) that videogames elicit when they allow players to customize the gaming experience. In a videogame setting, the user becomes more immersed in the media experience itself, compared with other types of media experiences (Nicovich 2005). In addition, customization encourages players to engage with a brand through much deeper, more meaningful experiences (Clifford-Marsh 2009). Therefore, the practical advice to marketers is to invest as necessary to allow players opportunities to interact and engage with the brand during the gaming experience, rather than simply seeing the brand’s logo displayed in various places within the game.
Reinforcing this suggestion is our finding that game customization is the lone significant predictor of recall of Mitsubishi, the integral brand. Although the Mitsubishi brand was prominently displayed throughout the game, such that all players had to use the brand for approximately nine minutes of game play, including practice, virtually no players recalled the brand unless they spent time interacting with it in the customization mode. Thus we might conclude that it was not worthwhile for Mitsubishi to invest to become the main racecar (integral brand), because recall improved only when players both interacted with and played the game with the integral brand, in support of Hang and Auty’s (2011) suggestion that inducing greater conceptual fluency through brand interaction or customization increases recall.
This reasoning suggests a potential link between in-game advertising effects and research on advergames. Mallinckrodt and Mizerski (2007) conceptualize interactive, brand-centered advergames as immersive opportunities that can reinforce brand images much better than traditional advertising. Their findings indicate that children who played an interactive advergame centered around a cereal brand demonstrated higher preference for the brand than for other cereals or other food types in general (Mallinckrodt and Mizerski 2007). More recent research (Lai and Huang 2011) indicates that advergames that require more cognitively demanding interactions also generate greater brand recall than games with lesser demands. Therefore, in line with our findings, the best investment in in-game brand communications appears to be efforts to increase integration and interaction opportunities (Wenner 2004), rather than pursuing simple placement efforts. This notion is even more compelling for brands considering only peripheral brand placements, which demonstrate very little effectiveness in our study.
Another important finding involves the impact of controller type. Our study participants demonstrated higher recall when they used a traditional controller rather than the steering wheel, which ostensibly is more technologically advanced and more naturally mapped (Skalski et al. 2007). It seems that the novelty of the wheel required players to allocate greater cognitive resources to controlling the car, which diminished the resources available to attend to in-game brand messages (Lang 2000, 2006; Lee and Faber 2007). This finding is not specific to integral or peripheral brand placements; the traditional controller enables better ad memory in general. Therefore, we offer a caution to marketers who believe an association with the latest technology or most advanced gaming systems is always a wise investment. It is likely that highly advanced technological systems, unfamiliar controls (even if they seem more natural), and ever-advancing game features can demand so many cognitive resources that players simply cannot attend to in-game brand messages.
The interaction effects of controller type and customization confirm that the traditional, familiar controller allows for the greatest amount of attention to in-game brand messages. Players who used the wheel and did not interact with the brand before the game were at a much greater disadvantage for remembering brand placements than any other treatment group. In contrast, players who used the traditional controller, even without being able to customize, displayed much higher recall. These effects were not specific to either the integral or peripheral brands, which is telling. It suggests that the amount of mere exposure to the integral brand through customization cannot explain our results; the interplay of customization and controller type affects recall for both integral and peripheral brands combined, regardless of the time spent interacting with the integral brand before the game. It also is interesting to note that variables related to players’ prior experience with the game and their in-lab gaming encounter could not account for differences in brand recall. Therefore, we are confident that our interpretation of the current findings, using LC4MP and processing fluency, is appropriate and goes beyond other, seemingly obvious explanations, all of which we have accounted for.
Overall, our findings indicate the compelling need for marketers and game creators to cooperate and advance each other’s interests. As gaming companies strive to put forth the best products, the brands that advertise in them-and thereby subsidize the cost of producing the games-must constantly ask the age-old question: Is advertising working? Many marketers covet positive brand associations, such that placing a logo or brand name within a popular game title seems like a good investment, but this study offers numerous caveats, mainly based on the level of brand integration within the game. In addition, simply throwing money at the latest technological advancements, regardless of their eventual popularity among gamers, is not a fail-safe method to increase brand effectiveness either. Similar to advertising concerns for any medium, it is paramount to keep returns on investment in mind when creating new media content in an effort to realize sales or market growth in the gaming industry. Advertisers and media/content providers must strive to understand each other’s needs when negotiating beneficial relationships.
As with any experiment, this study contains several limitations that are important to note. In particular, the current study included only one brief game-playing experience. Although we controlled for prior game experience, repeated exposure to the same brand messages while playing the same game many times might affect memory in different ways. However, repeated game play logically should increase the potential amount of player engagement and interaction with the integral brand through customization, which accordingly should produce positive effects, regardless of time spent playing. This claim is especially compelling when considering that customized game play has been linked to increased enjoyment and higher game repurchase intentions (Kwak et al. 2010). Thus, more marketing investment in integral, customizable brand placements seem likely to perpetuate a multilevel, positive cycle of brand outcomes.
Another limitation pertains to the selection of controller types. Although the new controller was the more naturally mapping of the two choices, no participants had ever used it before. It would be interesting to determine if the same controller-based effects are manifest for a new controller that is not completely novel to participants. Perhaps additional studies could use naturally mapping controllers with which participants have some experience, albeit still less than with traditional controllers. Absent the novelty of having to learn a brand new device completely from scratch, perhaps more cognitive resources could be allocated to the game, which might increase subsequent brand recall.
Finally, the use of other memory-based measures would be useful for clarifying the demonstrated relationships between game customization and control mechanism. Although free recall is appropriate for this study, as used in the vast majority of studies on videogame advertising, the inclusion of other ad- or brand-related measures might reveal if the two factors we study have impacts beyond retrieval measures. For example, both the LC4MP and processing fluency models espouse cue-based responses, such as recognition and aided recall, to gauge dimensions of the memory process that differ from those examined herein. Other, more affective measures also could determine if players might associate good feelings or moods with certain brands and/or in-game experiences. Although such insight would not address memory-based attributes, it could offer important implications for both game makers and marketers looking to advertise their brands within videogames.
References
Bailey, Rachel, Kevin Wise, and Paul Bolls (2009), “How Avatar Customizability Affects Children’s Arousal and Subjective Presence during Junk Food-Sponsored Online Video Games,” CyberPsychology & Behavior, 12, 277-83.
Balasubramanian, Siva K. (1994), “Beyond Advertising and Publicity: Hybrid Messages and Public Policy Issues,” Journal of Advertising, 23 (4), 29-46.
Chaney, Isabella M., Ku-Ho Lin, and James Chaney (2004), “The Effect of Billboards within the Gaming Environment,” Journal of Interactive Advertising, 5 (1), <http://www.jiad.org/article54>.
Chen, Jenova (2007), “Flow in Games (and Everything Else),” Communications of the ACM, 50 (4), 31-34.
Clifford-Marsh, Elizabeth (2009), “In-Game Advertising,” Revolution, (January), 46-47.
Gee, James Paul (2005), “Learning by Design: Good Video Games as Learning Machines,” E-Learning, 2, 5-16.
Guadiosi, John (2008), “Gaming,” Advertising Age, 79 (11), 38.
Gupta, Pola B., and Kenneth R. Lord (1998), “Product Placement in Movies: The Effect of Prominence and Mode on Audience Recall,” Journal of Current Issues and Research in Advertising, 20 (1), 47-59.
Hang, Haiming, and Susan Auty (2011), “Children Playing Branded Video Games: The Impact of Interactivity on Product Placement Effectiveness,” Journal of Consumer Psychology, 21 (1), 65-72.
Hartley, Matt (2007), “Why Is the Nintendo Wii So Successful?” <http://www.smarthouse.com.au/Gaming/Industry/F4U3N4C3?page=3> (accessed on October 20, 2008).
Ivory, James D. and Sriram Kalyanaraman (2007), “The Effect of Technological Advancement and Violent Content in Video Games on Players’ Feelings of Presence, Involvement, Physiological Arousal, and Aggression,” Journal of Communication, 57, 532-55.
Kahneman, Daniel (1973), Attention and Effort. Englewood Cliffs, NJ: Prentice Hall.
Karrh, James A. (1998), “Brand Placement: A Review,” Journal of Current Issues and Research in Advertising, 20 (2), 31-49.
—, Kathy Brittain McKee, and Carol J. Pardun (2003), “Practitioners’ Evolving Views on Product Placement Effectiveness,” Journal of Advertising Research, 43, 138-49.
Kwak, Dae Hee, Galen E. Clavio, Andrea N. Eagleman, and Ki Tak Kim (2010), “Exploring the Antecedents and Consequences of Personalizing Sport Video Game Experiences,” Sport Marketing Quarterly, 19 (4), 217-25.
Lai, Mengkuan and Ya-Shin Huang (2011), “Can Learning Theoretical Approaches Illuminate the Ways in which Advertising Games Effect Attitude, Recall, and Purchase Intention,” International Journal of Electronic Business Management, 9 (4), 368-80.
Lang, Annie (2000), “The Limited Capacity Model of Mediated Message Processing,” Journal of Communication, 50 (1), 46-70.
— (2006), “Using the Limited Capacity Model of Motivated Mediated Message Processing to Design Effective Cancer Communication Messages,” Journal of Communication, 56, S57-80.
—, Robert F. Potter, and Paul D. Bolls (1999), “Something for Nothing: Is Visual Encoding Automatic?” Media Psychology, 1, 145-63.
Lee, Angela Y. and Aparna A. Labroo (2004), “Effects of Conceptual and Perceptual Fluency on Brand Evaluation,” Journal of Marketing Research, 41 (2), 151-65.
Lee, Mira and Ronald J. Faber (2007), “Effects of Product Placement in On-Line Games and Brand Memory: A Perspective of The Limited-Capacity Model of Attention,” Journal of Advertising, 36 (4), 75-90.
Limperos, Anthony M., Mike G. Schmierbach, Andrew D. Kegerise, and Frank E. Dardis (2011), “Gaming Across Different Consoles: Exploring the Role of Control Scheme on Game Player Enjoyment,” Cyberpsychology, Behavior, and Social Networking, 14, 345-50.
Lin, James J. W., Henry Been Lirn Duh, Habib Abi-Rached, Donald E. Parker, and Thomas A. Furness (2002), “Effects of Field of View on Presence, Enjoyment, Memory, and Simulator Sickness in a Virtual Environment,” paper presented at the IEEE Virtual Reality Conference, March, Orlando, FL.
Lindley, Sian, James Le Couteur, and Nadia Bianchi-Berthouze (2008), “Stirring Up Experience Through Movement in Gameplay: Effects on Engagement and Social Behaviour,” Proceedings of SIGCHI Conference on Human Factors in Computing Systems, 511-14.
Lynch, John G. and Thomas K. Srull (1982), “Memory and Attentional Factors in Consumer Choice: Concepts and Research Methods,” Journal of Consumer Research, 9 (1), 18-37.
“Major League Gaming Research Reveals Competitive Video Gaming Audience as Lifestyle Leaders in Social Activity, Sponsor Empathy, and Media Consumption” (2009), PR Newswire, January 12.
Mallinckrodt, Victoria and Dick Mizerski (2007), “The Effects of Playing an Advergame on Young Children’s Perceptions, Preferences, and Requests,” Journal of Advertising, 36 (2), 87-100.
Nelson, Michelle R. (2002), “Recall of Brand Placements in Computer/Video Games,” Journal of Advertising Research, 2, 80-92.
—, Ronald A. Yaros, and Heejo Keum (2006), “Examining the Influence of Telepresence on Spectator and Player Processing of Real and Fictitious Brands in a Computer Game,” Journal of Advertising, 35 (4), 87-99.
Nicovich, Stefan G. (2005), “The Effect of Involvement on Ad Judgment in a Video Game Environment: The Mediating Role of Presence,” Journal of Interactive Advertising, 6 (1), <http://www.jiad.org/article67>.
Parker, Seb (2010), “2010 Year on Year Sales and Market Share Update to September 25th,” <http://www.vgchartz.com/article/82283/2010-year-on-year-sales-and-market-share-update-to-september-25th/> (accessed October 3, 2010).
Persky, Susan and Jim Blaskovich (2007), “Immersive Virtual Environments versus Traditional Platforms: Effects of Violent and Nonviolent Video Game Play,” Media Psychology, 10, 135-56.
Pinelle, David, Nelson Wong, and Tadeusz Stach (2008), “Heuristic Evaluation for Games: Usability Principles for Video Game Design,” Proceedings of SIGCHI Conference on Human Factors in Computing Systems, 1453-62.
Schneider, Lars-Peter and T. Bettina Cornwell (2005), “Cashing in on Crashes via Brand Placement in Computer Games: The Effects of Experience and Flow on Memory,” International Journal of Advertising, 24 (3), 321-43.
Skalski, Paul, Ryan L. Lange, Ron Tamborini, and Ashleigh K. Shelton (2007), “Mapping the Road to Fun: Natural Video Game Controllers, Presence, and Game Enjoyment,” paper presented at the annual meeting of the International Communication Association, May, San Francisco, CA.
Shapiro, Stewart (1999), “When an Ad’s Influence Is Beyond Our Conscious Control: Perceptual and Conceptual Fluency Effects Caused by Incidental Ad Exposure,” Journal of Consumer Research, 26 (1), 16-36.
Tamborini, Ron and Paul Skalski (2006), “The Role of Presence in the Experience of Electronic Games,” in Playing Video Games: Motives, Responses, and Consequences, Peter Vorderer and Jennings Bryant, eds. Mahwah, NJ: Lawrence Erlbaum, 225-40.
Tulving, Endel, and Daniel L. Schacter (1990), “Priming and Human Memory Systems,” Science, 247 (January), 301-306.
Walsh, Patrick, Yongjae Kim, and Stephen D. Ross (2008), “Brand Recall and Recognition: A Comparison of Television and Sport Video Games as Presentation Models,” Sport Marketing Quarterly, 17 (4), 201-208.
Wenner, Lawrence A. (2004), “On the Ethics of Product Placement in Media Entertainment,” Journal of Promotion Management, 10 (1/2), 101-32.
Whittlesea, Bruce W. A. (1993), “Illusions of Familiarity,” Journal of Experimental Psychology: Learning, Memory, and Cognition, 19 (6), 1235-53.
Yee, Nick (2006), “Motivations for Play in Online Games,” CyberPsychology & Behavior, 9, 772-75.
About the Authors
Frank E. Dardis (Ph.D., University of South Carolina) is an associate professor in the Department of Advertising and Public Relations in the College of Communications at Pennsylvania State University. His research focuses on the persuasive effects of strategic communications in both consumer-oriented and sociopolitical contexts. Email: [email protected]
Mike Schmierbach (Ph.D., University of Wisconsin-Madison) is an assistant professor in the College of Communications at Pennsylvania State University. Much of his research in media effects currently focuses on the impact of videogames. Email: [email protected]
Anthony M. Limperos (Ph.D., Pennsylvania State University) is an assistant professor in the School of Journalism and Telecommunications at the University of Kentucky. His research interests include media psychology and the uses and effects of video games and new communication technology. Email: [email protected]
